Modular RAG framework for experimentation.
The RAG Framework Python Package is designed to provide a convenient way to interact with various components of the RAG (Retrieval-Augmented Generation) framework. RAG is a powerful framework used for generating text-based responses, where users can select specific components like VectorDB, Language Model (LLM), Orchestrator, Langchain, or the Llama index, or simply use plain Python to generate text completions. The package allows users to select the desired components and obtain outputs from the RAG framework, including different methods of employing RAG.
To use this package, follow these steps:
-
Clone the Repository: Clone the RAG Framework Python Package repository from GitHub.
git clone https://github.com/SDcodehub/TryRAG.git cd rag-framework-python-package -
Create a Virtual Environment: It is recommended to create a virtual environment to manage dependencies.
python -m venv venv source venv/bin/activate # On Windows, use "venv\Scripts\activate"
-
Install Dependencies: Install the required dependencies from the
requirements.txtfile.pip install -r requirements.txt
-
API Key Configuration:
-
Create a
.envfile in the project directory.touch .env
-
Open the
.envfile and add your OpenAI API key as follows:OPENAI_API_KEY=your_api_key_here
-
- For more information on the RAG framework and its components, refer to the official RAG framework documentation.
- Additional methods for employing RAG can be explored based on the specific use case and requirements.
This package is distributed under the MIT License. See the LICENSE file for more details.
Note: This README provides a basic structure. You may need to update and refine it based on the specific implementation of your RAG framework and package details.